This comprehensive survey analyzes the use of large language models for data processing in IoT systems, highlighting integration challenges and opportunities.
The Internet of Things (IoT) has emerged as a technological cornerstone, enabling highly interconnected systems that integrate billions of diverse devices, with a combined data output estimated at hundreds of millions of terabytes per day. The rise of Large Language Models (LLMs) presents new opportunities for managing these devices, processing the data they generate, and addressing the technological and societal challenges posed by IoT ecosystems. However, the integration of LLMs into IoT also introduces significant challenges. This article offers the most comprehensive survey to date on the integration of LLMs into IoT ecosystems. We cover various technical layers, including device, software engineering, sensing, networking, data processing, privacy and security, and human interaction. We synthesize insights from over 300 articles, identify open research gaps, and highlight promising future directions, ranging from semantic reasoning and communication to the integration of quantum computing. By addressing the characteristics, challenges, and opportunities associated with LLM integration, this article serves as a foundational resource for researchers and practitioners aiming to enhance IoT functionalities with LLMs while navigating the associated complexities.
No takes yet. Share an insight, caveat, or question.
Sarhaddi et al. (2025) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: